Real-time traffic monitoring by fusing floating car data with stationary detector data

Houbraken, Maarten; Audenaert, Pieter; Colle, Didier; Pickavet, Mario; Scheerlinck, Karolien; Yperman, I; Logghe, Steven

doi:10.1109/mtits.2015.7223247

Cited by 6 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…-19. 3. 2021, Ostrava The authors have been dealing with the issue of floating vehicles for a long time.…”

Section: Gis Ostrava 2021 -Prostorové Služby Pro Smart City a Smart Regionmentioning

confidence: 99%

See 1 more Smart Citation

Sledování dopravního proudu na základě dat z plovoucích vozidel v GIS

Purkrábková¹,

Hrubeš²

2021

GIS Ostrava 2021 - Prostorové Služby Pro Smart City a Smart Region

View full text Add to dashboard Cite

Moderním trendem oblastí "Smart" je sběr dat k ovlivňování, řízení, optimalizaci nebo predikci obecných procesů, dopravy nevyjímaje. V České republice jsou od loňského roku k dispozici data modelu o rychlosti silničních dopravních proudů vycházející z pohybu tzv. plovoucích vozidel, tedy nový zdroj dat o dopravní situaci s celoplošným pokrytím. Tato data jsou publikována tedy nejen na tradičních hlavních komunikacích, ale také na komunikacích, které dosud nebyly a nejsou vybavené žádnou detekční technologií. Díky tomuto modelu je možné sledovat a analyzovat rychlost dopravního proudu a jeho vývoj v čase pro velkou část dopravní sítě celé ČR. Článek pojednává o využití těchto dat, jejich prostorovému významu a zpracování v GIS

show abstract

“…-19. 3. 2021, Ostrava The authors have been dealing with the issue of floating vehicles for a long time.…”

Section: Gis Ostrava 2021 -Prostorové Služby Pro Smart City a Smart Regionmentioning

confidence: 99%

“…In the vast majority of cases, these are pilot projects on the motorway. Most of these projects also used data for traffic management on arrivals to large metropolises, such as [3]. Similar data has already been processed using GIS, for example by research [4], which also uses data from taxi vehicles.…”

Section: State-of-the-artmentioning

confidence: 99%

Sledování dopravního proudu na základě dat z plovoucích vozidel v GIS

Purkrábková¹,

Hrubeš²

2021

GIS Ostrava 2021 - Prostorové Služby Pro Smart City a Smart Region

View full text Add to dashboard Cite

show abstract

“…Hence, significant efforts have been devoted towards development of accurate and fast automatic incident detection (AID) algorithms. Traditional AID algorithms rely on radar-based sensor data [3], loop detector data [4], probe vehicle data [5], or fusing multiple multiple streams [6], [7] to detect traffic incidents from data streams. State Departments of Transportation (DoTs) also install a large number of Closed Circuit Television (CCTV) cameras for surveillance tasks along freeways.…”

Section: Introductionmentioning

confidence: 99%

Freeway Traffic Incident Detection from Cameras: A Semi-Supervised Learning Approach

Chakraborty

Sharma

Hegde

2018

2018 21st International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

Early detection of incidents is a key step to reduce incident related congestion. State Department of Transportation (DoTs) usually install a large number of Close Circuit Television (CCTV) cameras in freeways for traffic surveillance. In this study, we used semi-supervised techniques to detect traffic incident trajectories from the cameras. Vehicle trajectories are identified from the cameras using state-ofthe-art deep learning based You Look Only Once (YOLOv3) classifier and Simple Online Realtime Tracking (SORT) is used for vehicle tracking. Our proposed approach for trajectory classification is based on semisupervised parameter estimation using maximum-likelihood (ML) estimation. The ML based Contrastive Pessimistic Likelihood Estimation (CPLE) attempts to identify incident trajectories from the normal trajectories. We compared the performance of CPLE algorithm to traditional semi-supervised techniques Self Learning and Label Spreading, and also to the classification based on the corresponding supervised algorithm. Results show that approximately 14% improvement in trajectory classification can be achieved using the proposed approach.

show abstract

“…Both FCD and loop data are used simultaneously in [28] to estimate the fundamental diagram. Similarly, Rahmani et al [29] fuse camera data with FCD, while Houbraken et al [30] fuse loop data with FCD to estimate traffic conditions. This paper focuses on the potential of FCD for dynamic traffic management, more specifically VSL to a.o.…”

Section: Introductionmentioning

confidence: 99%

Examining the potential of floating car data for dynamic traffic management

Houbraken

Logghe²,

Audenaert

et al. 2018

IET Intelligent Transport Systems

Self Cite

View full text Add to dashboard Cite

Traditional traffic monitoring systems are mostly based on road side equipment measuring traffic conditions throughout the day. With more and more GPS enabled connected devices, Floating Car Data (FCD) has become an interesting source of traffic information, requiring only a fraction of the road side equipment infrastructure investment. While FCD is commonly used to derive historic travel times on individual roads and to evaluate other traffic data and algorithms, it could also be used in traffic management systems directly. However, as live systems only capture a small percentage of all traffic, its use in live operating systems needs to be examined. In this paper, we investigate the potential of FCD to be used as input data for live automated traffic management systems. The FCD in this study is collected by a live country-wide FCD system in the Netherlands covering 6-8% of all vehicles. The (anonymised) data is first compared to available road side measurements to show the current quality of FCD. It is then used in a dynamic speed management system and compared to the installed system on the studied highway. Results indicate the FCD setup can approximate the installed system, showing the feasibility of a live system.

show abstract

Real-time traffic monitoring by fusing floating car data with stationary detector data

Cited by 6 publications

References 14 publications

Sledování dopravního proudu na základě dat z plovoucích vozidel v GIS

Sledování dopravního proudu na základě dat z plovoucích vozidel v GIS

Freeway Traffic Incident Detection from Cameras: A Semi-Supervised Learning Approach

Examining the potential of floating car data for dynamic traffic management

Contact Info

Product

Resources

About